Upload folder using huggingface_hub

README.md

@@ -33,7 +33,9 @@ vllm serve $model_id \
 ```bash
 # Multi-token prediction is supported
 model_id=tiny-random/qwen3.5-moe
-python3 -m sglang.launch_server   --model-path $model_id   --tp-size 2 \
+python3 -m sglang.launch_server \
+  --model-path $model_id \
+  --tp-size 2 \
   --tool-call-parser qwen3_coder  \
   --reasoning-parser qwen3 \
   --speculative-algo NEXTN \
@@ -84,7 +86,7 @@ inputs = processor.apply_chat_template(
 ).to(model.device)
 
 generated_ids = model.generate(**inputs, max_new_tokens=32)
-output_text = processor.batch_decode(generated_ids[0])
+output_text = processor.batch_decode(generated_ids[0][inputs['input_ids'].shape[1]:], skip_special_tokens=True)
 print(output_text)
 ```
 

model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0aac8f1e5ef91b3c9c68c394611de2333b4e76fd229ef13864793d5826fa024c
+oid sha256:3a68280ca252dadbb9875aa7455a9def12207fc29aa7bf34f10f578a4e4102cb
 size 10057952